Using Business Intelligence to Enhance Practice Management

In an environment where radiologists are consistently being asked to do more with less, a simple increase in production is no longer a viable remedy to a problem. Practice managers and physician owners need to work smarter, instead of just harder. Using business intelligence data for predictive analytics, they are beginning to run their practices more efficiently, but also more effectively, challenging the status quo and using data to affect change. Pat Epting, executive director of Radiology Associates of Clearwater (RAC) and former Radiology Business Management Association President, and Jana Landreth, CPA, MBA, director of practice management with Zotec Partners, share how they used business intelligence data to create a practice management model that works for RAC. The model has had a significant impact on the day-to-day operations, physician scheduling processes, relationships with customers, and has also contributed to the overall job satisfaction of the physicians.

If you do what you’ve always done…

“It was an evolution of the practice’s needs over time that really set us out on this journey,” recalls Epting. “During a period of growth, we took on more physician shareholders as our hospital systems customers consolidated. After some time, we were then facing the opposite situation wherein a number of our radiologists were retiring, and the remaining shareholders asked me, “How do we know if we need to replace them? Can you look into this? We’d like your advice.”

With hundreds of thousands of transactional data points at their disposal, Epting and Landreth used business intelligence to illustrate what services were required of the practice versus when the practice was producing them and created a predictive analytics model for optimizing staffing, the physician schedule and workflow management.

“We aggregated lines and lines of data to summarily describe what was happening in our business. Then, we used the analysis to advise our physicians and help them confidently manage their work day,” Epting continued.

“An evolution was also taking place among practice leaders,” Landreth added, “accepting and embracing the use of business intelligence to help manage what they had historically managed almost completely through experience and feelings.”

The result was a practice management model that predicts and monitors staffing, scheduling and workflow for the practice, as well as a clear understanding of the business as a whole, creating a “health check” for present operations and a solid foundation for future planning.

Looking beyond an increase in volume

For radiologists, who more than likely spent most of their young adult lives training to become doctors, not running businesses, the task of mining data to inform a decision is a daunting one. Landreth and Epting started their project by looking at the resources the practice was providing alongside the work that was being presented to the practice, and then analyzing whether or not the two were complimentary.

“Radiology groups are called upon more and more to use business intelligence to effectively manage resources and prove that effectiveness to customers,” Landreth counsels. There are only a certain number of hours in a day, therefore practice managers and physician owners have to look at the options the data is offering as quality evidence they can use to institute effective changes that will positively impact the business, rather than asking radiologists to increase their workload.

According to Landreth and Epting, the data used to create their model can typically be found in the scheduling software, the financial statements, the hospital information system, and the billing system. The challenge lies in that the information exists in organizational silos, isolated from the other information and rarely looked at holistically. “We married this information – the time, date and substance of the patient encounter performed by an identified physician on a specified shift with a certain financial outcome – to show a broader, aggregated perspective on the countless transactions taking place every day,” Landreth explains.

Practical applications

Many practices schedule physicians and staff based on a legacy model, referring to “the way it has always been done.” Radiologists typically complete a rotation in their practice schedules, so the distribution of work is equitable, or is perceived as such. When the industry transitioned to a PACS environment, radiologists moved from reading the film that was put in front of them to downloading images and reading from any location. Meanwhile practices have an opportunity to move to a relative value unit (RVU)-based system that quantifies work intensity, rather than using volume alone to create schedules or pay-for-performance criteria.

“Using transactional data, we can make adjustments down to the daily shift level,” says Epting. “One small change can make a big difference, whether it is altering shift start and end times, or staffing a site based on the individual performance of the physician.” Though the identifying data is removed when the aggregate physician performance data is presented to the group, Epting can see how effective each radiologist’s schedule may be for him or her, and changes may be implemented based on that information.

“We know everyone wants to work from 8 a.m. to 5 p.m., but that may not be the best scheduling option for our stakeholders. We use the model to see when work is ready to be interpreted versus staffing when we prefer to be there. We also know, for example, that our hospital customer slows down between 11 am and 1 pm to cycle through a lunch schedule. It may be more effective for our radiologist to assist another facility during that time, or provide specialty reads where there is a need, such as at the breast imaging center,” explains Epting. “We use the model to create each shift with a similar amount of work intensity (procedure volume and work RVUs) and then it doesn’t become as important to move people through a rotation, just to feed the perception of equivalent workloads for everyone. A benefit we’ve discovered as a result of that, is that it might be best for Dr. A to work at site X versus site Y. Having the same radiologist on staff at site X improves the service to the hospital customer and improves the consistency of the reads there. As well, Dr. A is able to forge a relationship with the hospital customer there, improving the overall service from our practice to our customers. Shift modeling creates equitable work, and in most cases, a less stressful workday for our physicians.”

Sharing best practices

Each and every radiology practice is different, but there are elements of this model that can be used ubiquitously; and there is certainly a need. In creating the model, Epting and Landreth moved beyond simply using resources more efficiently and effectively; they embraced the challenge to enhance the practice by using business intelligence data to advise shareholders on key decisions.

Epting and Landreth began this journey more than a half dozen years ago. With the knowledge and experience they’ve learned, other practices could generate their own model within a few months of imputing historical data. Landreth has shared the model with many of the practices with which she works, and many have adopted the practices that work for them, and Landreth and Epting have presented the information on multiple occasions, including during the RBMA Summit earlier this year.

“It really depends on the culture of the practice,” says Landreth. “It’s the right and privilege of physician business owners to accept or decline our suggestions. Ultimately every practice chooses the management model that works for them.”